CNN-Based Boat Detection Model for Alert System Using Surveillance Video Camera
Tatsuhiro Akiyama, Yosuke Kobayashi, Jay Junichi Kishigami, Kenji Muto · 2018
In Tokyo, various boats pass through the canal on the bayside. The loud sound created by these boats may cause some stress to the residents in that area. We propose a boat detection model based on convolutional neural networks (CNNs) using VGG19 that is trained using several types of boat pictures. Our proposed model aims to detect the type of boat passing through the canal using images obtained from the surveillance video camera. We finally achieve a practical result as F1-score of 0.70 by the proposed model.